
***DO FILE for "Can Sunlight Disperse Mistrust? A Meta-Analysis of the Effect of Transparency on Citizens’ Trust in Government"***


*using data “Study level averagez”*
*Figure 2——*
*Forest plot*
metan averagez averagese, random lcols(citation)


*using data “Meta-regression data”*
*Figure 3——*
*Funnel plot on effect size level*
confunnel fishersz se, xtitle (effect size) ytitle(Standard error of effect size)

*Publication bias test*
metabias fishersz se, egger
metabias fishersz se, begg


*Table 3——*
*Correlation Matrix*
pwcorr fishersz computermediatedtransparency decisionmakingtransparency hightransparency lowtransparency positivetransparency negativetransparency benevolence competence honesty experimental us journal, sig star(0.05) 


*Table 4——*
*Model 1*
metareg fishersz computermediatedtransparency benevolence competence honesty experimental us journal, wsse(se)
gen totalvarm1=var + 0.03239
gen resterrm1=sqrt(totalvarm1)
gen weight1re=(1/totalvarm1)
gen x0m1=1/resterrm1
gen Ywml=fishersz/resterrm1
gen computermediatedtransparency1=computermediatedtransparency/resterrm1
gen benevolence1=benevolence/resterrm1
gen competence1=competence/resterrm1
gen honesty1=honesty/resterrm1
gen experimental1=experimental/resterrm1
gen us1=us/resterrm1
gen journal1=journal/resterrm1
*CRVE*
reg Ywml computermediatedtransparency1 benevolence1 competence1 honesty1 experimental1 us1 journal1 x0m1, noc vce(cluster studyid)
*GEE*
xtset studyid
xtgee Ywml computermediatedtransparency1 benevolence1 competence1 honesty1 experimental1 us1 journal1 x0m1, nocon family(normal) link(identity) corr(exchangeable) vce(robust)

*Model 2* 
metareg fishersz decisionmakingtransparency benevolence competence honesty experimental us journal, wsse(se)
gen totalvarm1=var + 0.03269
gen resterrm1=sqrt(totalvarm1)
gen weight1re=(1/totalvarm1)
gen x0m1=1/resterrm1
gen Ywml=fishersz/resterrm1
gen decisionmakingtransparency1=decisionmakingtransparency/resterrm1
gen benevolence1=benevolence/resterrm1
gen competence1=competence/resterrm1
gen honesty1=honesty/resterrm1
gen sample1=sample/resterrm1
gen experimental1=experimental/resterrm1
gen us1=us/resterrm1
gen journal1=journal/resterrm1
*CRVE*
reg Ywml decisionmakingtransparency1 benevolence1 competence1 honesty1 experimental1 us1 journal1 x0m1, noc vce(cluster studyid)
*GEE*
xtset studyid
xtgee Ywml decisionmakingtransparency1 benevolence1 competence1 honesty1 experimental1 us1 journal1 x0m1, nocon family(normal) link(identity) corr(exchangeable) vce(robust)

*Model 3* 
metareg fishersz hightransparency lowtransparency benevolence competence honesty  experimental us journal, wsse(se)
gen totalvarm1=var + 0.03341
gen resterrm1=sqrt(totalvarm1)
gen weight1re=(1/totalvarm1)
gen x0m1=1/resterrm1
gen Ywml=fishersz/resterrm1
gen hightransparency1=hightransparency/resterrm1
gen lowtransparency1=lowtransparency/resterrm1
gen benevolence1=benevolence/resterrm1
gen competence1=competence/resterrm1
gen honesty1=honesty/resterrm1
gen experimental1=experimental/resterrm1
gen us1=us/resterrm1
gen journal1=journal/resterrm1
*CRVE*
reg Ywml hightransparency1 lowtransparency1 benevolence1 competence1 honesty1 experimental1 us1 journal1 x0m1, noc vce(cluster studyid)
*GEE*
xtset studyid
xtgee Ywml hightransparency1 lowtransparency1 benevolence1 competence1 honesty1 experimental1 us1 journal1 x0m1, nocon family(normal) link(identity) corr(exchangeable) vce(robust)

*Model 4* 
metareg fishersz positivetransparency negativetransparency benevolence competence honesty experimental us journal, wsse(se)
gen totalvarm1=var + 0.03311
gen resterrm1=sqrt(totalvarm1)
gen weight1re=(1/totalvarm1 )
gen x0m1=1/resterrm1
gen Ywml=fishersz/resterrm1
gen positivetransparency1=positivetransparency/resterrm1
gen negativetransparency1=negativetransparency/resterrm1
gen benevolence1=benevolence/resterrm1
gen competence1=competence/resterrm1
gen honesty1=honesty/resterrm1
gen experimental1=experimental/resterrm1
gen us1=us/resterrm1
gen journal1=journal/resterrm1
*CRVE*
reg Ywml positivetransparency1 negativetransparency1 benevolence1 competence1 honesty1 experimental1 us1 journal1 x0m1, noc vce(cluster studyid)
*GEE*
xtset studyid
xtgee Ywml positivetransparency1 negativetransparency1 benevolence1 competence1 honesty1  experimental1 us1 journal1 x0m1, nocon family(normal) link(identity) corr(exchangeable) vce(robust)


*Appendix 3——*
*Model 1*
winsor2 fishersz se, replace cuts(1 99)
metareg fishersz computermediatedtransparency benevolence competence honesty experimental us journal, wsse(se)
gen totalvarm1=var + 0.02986
gen resterrm1=sqrt(totalvarm1)
gen weight1re=(1/totalvarm1 )
gen x0m1=1/resterrm1
gen Ywml=fishersz/resterrm1
gen computermediatedtransparency1=computermediatedtransparency/resterrm1
gen benevolence1=benevolence/resterrm1
gen competence1=competence/resterrm1
gen honesty1=honesty/resterrm1
gen experimental1=experimental/resterrm1
gen us1=us/resterrm1
gen journal1=journal/resterrm1
*CRVE*
reg Ywml computermediatedtransparency1 benevolence1 competence1 honesty1 experimental1 us1 journal1 x0m1, noc vce(cluster studyid)
*GEE*
xtset studyid
xtgee Ywml computermediatedtransparency1 benevolence1 competence1 honesty1 experimental1 us1 journal1 x0m1, nocon family(normal) link(identity) corr(exchangeable) vce(robust)

*Model 2*
winsor2 fishersz se, replace cuts(1 99)
metareg fishersz decisionmakingtransparency benevolence competence honesty experimental us journal, wsse(se)
gen totalvarm1=var + 0.03049
gen resterrm1=sqrt(totalvarm1)
gen weight1re=(1/totalvarm1 )
gen x0m1=1/resterrm1
gen Ywml=fishersz/resterrm1
gen decisionmakingtransparency1=decisionmakingtransparency/resterrm1
gen benevolence1=benevolence/resterrm1
gen competence1=competence/resterrm1
gen honesty1=honesty/resterrm1
gen experimental1=experimental/resterrm1
gen us1=us/resterrm1
gen journal1=journal/resterrm1
*CRVE*
reg Ywml decisionmakingtransparency1 benevolence1 competence1 honesty1 experimental1 us1 journal1 x0m1, noc vce(cluster studyid)
*GEE*
xtset studyid
xtgee Ywml decisionmakingtransparency1 benevolence1 competence1 honesty1 experimental1 us1 journal1 x0m1, nocon family(normal) link(identity) corr(exchangeable) vce(robust)

*Model 3* 
winsor2 fishersz se, replace cuts(1 99)
metareg fishersz hightransparency lowtransparency benevolence competence honesty  experimental us journal, wsse(se)
gen totalvarm1=var + 0.0309
gen resterrm1=sqrt(totalvarm1)
gen weight1re=(1/totalvarm1)
gen x0m1=1/resterrm1
gen Ywml= fishersz/resterrm1
gen hightransparency1=hightransparency/resterrm1
gen lowtransparency1=lowtransparency/resterrm1
gen benevolence1=benevolence/resterrm1
gen competence1=competence/resterrm1
gen honesty1=honesty/resterrm1
gen experimental1=experimental/resterrm1
gen us1=us/resterrm1
gen journal1=journal/resterrm1
*CRVE*
reg Ywml hightransparency1 lowtransparency1 benevolence1 competence1 honesty1 experimental1 us1 journal1 x0m1, noc vce(cluster studyid)
*GEE*
xtset studyid
xtgee Ywml hightransparency1 lowtransparency1 benevolence1 competence1 honesty1 experimental1 us1 journal1 x0m1, nocon family(normal) link(identity) corr(exchangeable) vce(robust)

*Model 4* 
winsor2 fishersz se, replace cuts(1 99)
metareg fishersz positivetransparency negativetransparency benevolence competence honesty experimental us journal, wsse(se)
gen totalvarm1=var + 0.03067
gen resterrm1=sqrt(totalvarm1)
gen weight1re=(1/totalvarm1 )
gen x0m1=1/resterrm1
gen Ywml=fishersz/resterrm1
gen positivetransparency1=positivetransparency/resterrm1
gen negativetransparency1=negativetransparency/resterrm1
gen benevolence1=benevolence/resterrm1
gen competence1=competence/resterrm1
gen honesty1=honesty/resterrm1
gen experimental1=experimental/resterrm1
gen us1=us/resterrm1
gen journal1=journal/resterrm1
*CRVE*
reg Ywml positivetransparency1 negativetransparency1 benevolence1 competence1 honesty1 experimental1 us1 journal1 x0m1, noc vce(cluster studyid)
*GEE*
xtset studyid
xtgee Ywml positivetransparency1 negativetransparency1 benevolence1 competence1 honesty1  experimental1 us1 journal1 x0m1, nocon family(normal) link(identity) corr(exchangeable) vce(robust)

